Sentimental Analysis Applications and Approaches during COVID-19: A Survey

Validating part of the social media infodemic listening conceptual framework using structural equation modelling

“… 36 , 37 This could lead to even fewer representative samples since the majority of Twitter users do not include demographic information in their profiles. Additionally, there are other methods for calculating sentiments, 8 , 9 although VADER sentiment analysis has commonly been used. 23 The data transformation via the natural logarithm also limited the data quality due to information loss.…”

“… 1 , 2 , 3 , 4 , 5 , 6 In addition to typical online questionnaires or qualitative analysis, researchers have applied machine learning (ML) or artificial intelligence (AI) techniques to investigate and better understand public discourse and sentiments and infer people's COVID-19 vaccine intentions. 7 , 8 , 9 The World Health Organization coined the term “social listening” to describe such activities and deployed its Early AI-Supported Response with Social Listening (EARS) platform during the pandemic. 10 …”

“…Social listening studies have adopted existing theories from health behaviours, communication, and behavioural sciences. 7 , 8 , 9 , 10 However, these models have limitations, and they do not truly reflect current complex information ecosystems. The literature includes various social listening studies that investigate the impacts of exposure to information circulated on social media platforms on individuals' intentions or behaviours in relation to COVID-19 vaccination.…”